There have been no systematic studies to date to determine whether the majority of identified avian and mammalian oncogenes have homologues in fish species. The first goal of the proposed research is to identify proto-oncogenes in genomic DNA prepared from fishes of the genus Xiphophorus (Poeciliidae) based on nucleotide sequence homologies with viral and mammalian oncogene probes. Restriction fragment length polymorphisms (RFLPs) in fish proto-oncogenes identified as a result will then be used in mapping Xiphophorus proto-oncogenes in backcross hybrids by testing for linkage to polymorphic enzyme loci already assigned to multipoint linkage groups. In Xiphophorus, a variety of genetically controlled tumors are produced in congeneric hybrids; therefore, mapping of fish proto-oncogenes in this genus will provide an important body of data which should be useful in correlating phenotypic characteristics (e.g. tumor severity, benign vs. malignant growth) in both F1 and backcross hybrids with inheritance patterns of specific proto-oncogene alleles. In the Xiphophorus hybrid melanoma system, levels of proto-oncogene expression (i.e., for c-src, c-myc, and c-myb) will be assessed by comparison of polyadenylated RNA signals obtained from melanotic and normal tissues. We will also use this experimental tumor model to assess any changes in gene expression which occur in the promotion of genetically benign melanomas to malignancy by treatment with hormones and known tumor promoters. Since there are a large number of identified fish tumors, the data obtained from these approaches should serve as a stimulus for the experimental determination of the involvement of oncogenes in other fish tumor models; some of these experimental tumor systems may prove to be important models for human neoplasia.